Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    instances of the word "order."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    "order to"

    np_acts-logits-general · gemini-2.0-flash

    phrases signaling purpose

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    DebuggerStep
    -0.78
    OGND
    -0.68
    fromnode
    -0.67
    DockStyle
    -0.63
     fieldNum
    -0.63
    SerializedSize
    -0.63
    StructEnd
    -0.61
    ьаж
    -0.60
    XMLSchema
    -0.59
    MessageTagHelper
    -0.58
    POSITIVE LOGITS
     words
    0.60
     maximize
    0.58
     Words
    0.55
     successfully
    0.53
    words
    0.53
     avoid
    0.52
     WORDS
    0.52
     maximise
    0.51
    Words
    0.51
    CloseOperation
    0.51
    Act Density 0.052%

    No Known Activations